simple grep expressions

Question

I have a few grep expressions that aren't working as I intended and I cannot figure out as to why.

The first expression is to find lines that start and end with the same character. This is the expression I am using...

grep -E '(.).*\1$' input

The second expression is very similar as it is to find lines that have the same 2nd character and second to last character...

grep -E '(.)(.).*\2.$' input

The last expression is to find lines with only a single word and end with a punctuation...

grep -E '(\w){1}.*[[:punct:]]$' input

I don't understand why these are not working, am i doing something terribly wrong or just a simple mistake?

Answer 1

You left off the start-of-line anchor (caret symbol). Here are modified commands which only match entire lines (and won't match substrings):

grep -E '^(.).*\1$' input

grep -E '^(.)(.).*\2.$' input

grep -E '^(\w){1}.*[[:punct:]]$' input

You'll probably also want to modify your second and third commands.

The second command doesn't require two capture groups. You can do it with just one:

grep -E '^.(.).*\1.$' input

The third command is incorrect: it will match strings with multiple words. A corrected version might be:

grep -E '^\w+[[:punct:]]$' input